Cell killing at therapeutic doses of ionizing radiation appears to be due principally to lesions produced by multiple radical formation within a small volume of DNA. We propose to develop model systems for studying such lesions, and to use these methods to develop new strategies for modification of radiation sensitivity for clinical application in the treatment of cancer. The methods to be developed will be refinements of existing techniques (neutral filter elution of DNA, electrophoretic analysis of damaged plasmids). Enzymatic base excision will be used to allow assessment of differential modifiability of damage to the bases in DNA and damage to the sugar backbone. Our experiments will focus on the rapid radiochemical reactions which determine the initial damage produced by radiation and which precede enzymatic modification of the damage. Our experimental approach is to vary parameters which we propose to be related to radiosensitivity, in tightly controlled systems of varying complexity. The simpler systems (DNA in solution, plasmids, isolated nuclei) will allow more precise control of variables, and more precise analysis of chemical endpoints. Intact cells will be used to confirm the biological significance of the results. Modifying agents to be studied include various thiols, O2, misonidazole, SR-2508, metal ions, and thiol-containing proteins. Mechanisms which will be examined include competition between protectors and sensitizers for reaction with DNA radicals, metal ion catalysis of multiple radical reactions, and charge dispersion by chromatin semiconductivity.